New Study: Artificial Intelligence Outperforms Clinical Risk Models in Predicting Breast Cancer Risk

A new study suggests another way Artificial intelligence enters the medical field and improving current practices for predicting breast cancer risk.

The study, published Tuesday in Radiology, a peer-reviewed journal, found that artificial intelligence algorithms outperformed standard clinical risk models for predicting risk over a five-year period. breast cancer.

Risk models such as the Breast Cancer Surveillance Consortium (BCSC) Clinical Risk Group, which use self-reported and other patient information including age, family history, and more, are commonly used to calculate a woman’s risk of developing breast cancer.

“Clinical risk models rely on gathering information from multiple sources, which are not always available or collated,” said lead researcher Dr. Vignesh A. Arasu, a research scientist and practitioner of radiology at Kaiser Permanente of Northern California, in a news release. “Recent advances in deep learning AI give us the ability to extract hundreds of thousands of additional mammogram features.”

In this retrospective study, thousands of mammograms were analyzed, and breast cancer risk scores were generated over five years by five artificial intelligence algorithms. These scores were then compared with each other and the BCSC clinical risk score.

“The five AI algorithms performed better than the BCSC risk model for predicting breast cancer risk at ages 0 to 5 years,” said Araso. “This robust predictive performance over a five-year period suggests that AI identifies missed cancers and characteristics of breast tissue that help predict future cancer development.”

While some institutions are already using AI to help detect cancer in mammograms, these findings suggest that AI could be a vital tool for helping patients’ risk scores in the future – which takes seconds to produce AI, according to the release.

“Artificial intelligence for predicting cancer risk gives us the opportunity to personalize the treatment of each woman, which is not available systematically,” said Araso. “This is a tool that can help us provide precise, personalized medicine on a national scale.”

The story continues

All your mammogram and breast cancer questions answered by medical experts

Ronnie contemplates D-Day

PGA and LIV Golf to merge and form a new entity

Why commercial real estate creates economic problems